Pyramidal Person Re-IDentification via Multi-Loss Dynamic Training

TL;DR

This paper introduces a pyramid model for person re-identification that works effectively without precise bounding boxes and employs a dynamic multi-loss training scheme, achieving state-of-the-art results.

Contribution

The paper proposes a novel coarse-to-fine pyramid model and a dynamic training scheme that together improve re-identification accuracy without relying on accurate bounding boxes.

Findings

Achieves state-of-the-art results on three datasets.

Exceeds current best method by 9.5% on CUHK03.

Effectively handles unaligned image pairs.

Abstract

Most existing Re-IDentification (Re-ID) methods are highly dependent on precise bounding boxes that enable images to be aligned with each other. However, due to the challenging practical scenarios, current detection models often produce inaccurate bounding boxes, which inevitably degenerate the performance of existing Re-ID algorithms. In this paper, we propose a novel coarse-to-fine pyramid model to relax the need of bounding boxes, which not only incorporates local and global information, but also integrates the gradual cues between them. The pyramid model is able to match at different scales and then search for the correct image of the same identity, even when the image pairs are not aligned. In addition, in order to learn discriminative identity representation, we explore a dynamic training scheme to seamlessly unify two losses and extract appropriate shared information between them. Experimental results clearly demonstrate that the proposed method achieves the state-of-the-art results on three datasets. Especially, our approach exceeds the current best method by 9.5% on the most challenging CUHK03 dataset.